Meiotic recombination serves to diversify the human genome, and this means of generating novel haplotypes in the MHC is likely to have significant consequences on the immune response to foreign material such as infectious pathogens and tumors. Efforts toward characterizing recombination within the MHC using sperm as a source of recombinant chromosomes are near completion. To date, over 20,000 sperm from 12 donors were successfully genotyped for two short tandem repeats (STRs) flanking the MHC. Six hundred recombinant sperm were identified, of which 346 carried a crossover within the MHC (3.3 Mb), and 331 were further localized within the MHC class I, II, and III regions. Twenty-four new and twelve previously published STRs were used to fine-map 273 of the MHC recombinants within intervals as small as 14 kb and no larger than 357 kb. Efforts are underway to assess the level of heterogeneity in recombination rates among individuals as well as detect the presence of segregation distortion. Studies that take into account both recombination fractions and linkage disequilibrium (ld) may provide meaning to conservation of the MHC linkage group, whether selected or coincidental. Ld is the nonrandom association of alleles at two linked loci. Strong ld across the MHC exists, particularly among alleles of specific multi-locus haplotypes and between particular genes within the complex. Ld is a measurement of historical events that may reflect selective pressures in some cases, whereas studies of recombinant chromosomes essentially measure real-time events, which are unlikely to reflect the protracted process of selection. For example, a reasonable conclusion from the observation of a hotspot for recombination in the region between a pair of genes that are in strong ld is that there has been selection for haplotypes composed of alleles at those or neighboring genes. Thus, analysis of ld statistics in the context of recombination data (i.e. genetic distance) could be a powerful tool for identification of potential selective pressures resulting in retention of certain haplotypes. We have completed molecular typing of 30 markers within and flanking the MHC. Preliminary analysis of ld between pairs of these 30 markers has indicated that, in general, global disequilibrium (Wn) values decrease with increasing physical map distance. Further, the range of Wn values for pairs of loci separated by about 100 Kb is exceedingly wide as is clearly illustrated in the HLA class II region. The data conform to the concept that a monotonic relationship between disequilibrium and distance does not necessarily occur in small genomic segments where factors such as drift, mutation, and admixture may overcome the effect of recombination. We have identified a "hotspot" for recombination in the 21 Kb region separating TAP1 from TAP2 and a detailed analysis of linkage disequilibrium between eight sites of variation within this region is currently being investigated. Another notable feature of the data is that significant levels of ld exist between many pairs of markers that are separated by more than 1Mb. Further analyses of these data are ongoing.